www.impactjournals.com/oncotarget/ Oncotarget, 2017, Vol. 8, (No.

16), pp: 27440-27453

Research Paper

Decreased ATF4 expression as a mechanism of acquired

resistance to long-term amino acid limitation in cancer cells
Florent Mesclon1, Sarah Lambert-Langlais2, Valérie Carraro1, Laurent Parry1,
Isabelle Hainault3, Céline Jousse1, Anne-Catherine Maurin1, Alain Bruhat1, Pierre
Fafournoux1, Julien Averous1
1
Université Clermont Auvergne, INRA, UNH, Unité de Nutrition Humaine, CRNH Auvergne, F-63000 Clermont-Ferrand, France
2
Department of Medical Biochemistry and Molecular Biology, CHU de Clermont-Ferrand, 63003 Clermont-Ferrand Cedex 1,
France
3
Institute of Cardiometabolism and Nutrition, Université Pierre et Marie Curie, INSERM, UMR S1138, Centre de Recherche
des Cordeliers, 75006 Paris, France
Correspondence to: Julien Averous, email: julien.averous@clermont.inra.fr
Pierre Fafournoux, email: pierre.fafournoux@clermont.inra.fr
Keywords: ATF4, amino acids, apoptosis, cancer, resistance
Received: July 22, 2016     Accepted: February 12, 2017     Published: March 02, 2017
Copyright: Mesclon et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License
(CC-BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source
are credited

ABSTRACT

The uncontrolled growth of tumor can lead to the formation of area deprived in
nutrients. Due to their high genetic instability, tumor cells can adapt and develop
resistance to this pro-apoptotic environment. Among the resistance mechanisms,
those involved in the resistance to long-term amino acid restriction are not
elucidated. A long-term amino acid restriction is particularly deleterious since nine
of them cannot be synthetized by the cells. In order to determine how cancer cells
face a long-term amino acid deprivation, we developed a cell model selected for its
capacity to resist a long-term amino acid limitation. We exerted a selection pressure
on mouse embryonic fibroblast to isolate clones able to survive with low amino acid
concentration. The study of several clones revealed an alteration of the eiF2α/ATF4
pathway. Compared to the parental cells, the clones exhibited a decreased expression
of the transcription factor ATF4 and its target genes. Likewise, the knock-down of
ATF4 in parental cells renders them resistant to amino acid deprivation. Moreover,
this association between a low level of ATF4 protein and the resistance to amino acid
deprivation was also observed in the cancer cell line BxPC-3. This resistance was
abolished when ATF4 was overexpressed. Therefore, decreasing ATF4 expression
may be one important mechanism for cancer cells to survive under prolonged amino
acid deprivation.

INTRODUCTION
Tumor cells are characterized by an uncontrolled
proliferation rate. Such anarchic development of the tumor
mass may be accompanied by an inadequate vascularization
of the tumors [1]. As a consequence, nutrients and oxygen
may not be provided sufficiently and could contribute to
growth arrest and necrosis of the tumor [2]. However, tumor
cells exhibit high genetic instability which can lead to the
development of resistance mechanisms to nutrients and
oxygen deprived environment [3]. Mechanisms concerning
lack of glucose and oxygen have been extensively studied [4,
5]. In contrast, less is known on how tumor cells can adapt to
the lack of lipids or amino acids. Since several years, there
is an increasing interest for the role of the non-essential
amino acids (glutamine, serine, glycine,..) as precursors
of intermediate metabolites in the context of cancer [6, 7].
However, it is unclear how tumor cells can face a global
amino acid restriction since nine essential amino acids
cannot be synthesized by any mammalian cell. It remains
puzzling that some tumor cells facing such a deficit are able
not only to survive, but also to maintain minimal growth.

www.impactjournals.com/oncotarget 27440 Oncotarget

 In contrast, it is established that in non-tumor cells the
withdrawal of one essential amino acid can promote growth
arrest and ultimately apoptosis if this condition persists [8].
Two pathways are well known to be controlled
by amino acid availability, GCN2 (General Control
Nonderepressible 2) and mTORC1 (mammalian Target
Of Rapamycin Complex 1). GNC2 is a kinase activated
under amino acid starvation [9]. The activation of GCN2
induces its auto-phosphorylation that contributes to the
efficient phosphorylation of its substrate, the translation
initiation factor eIF2α (eukaryotic Initiation Factor 2) [10].
This event leads to a global inhibition of protein synthesis
but also, derepresses the translation of specific mRNAs
possessing uORF (upstream Open Reading Frame) in
their 5’UTR such as the one encoding ATF4 (Activating
Transcription Factor 4) [11]. Expression of this transcription
factor leads to the expression of a subset of specific genes
involved in the adaptation to the lack of amino acids [12],
these genes being either pro-survival or pro-apoptotic. The
multi-protein complex mTORC1 is a major regulator of cell
growth, which coordinates anabolic and catabolic processes
according to nutrients and energy environment. The activity
of this complex relies on the protein mTOR which has been
initially identified as the kinase targeting S6K1 (ribosomal
protein S6 Kinase 1) and 4E-BP1(eIF4E-Binding Protein 1)
two proteins involved in the control of protein synthesis [13].
It is established that mTORC1 activity may be
induced in tumors as negative regulators of its activity are
often found to be mutated in cancer cells [14]. A constant
and uncontrolled activation of mTORC1 can contribute
significantly to the tumorigenesis through the control of
different cell functions such as translation or autophagy
[15]. Moreover, the overexpression of amino acid
transporters that contributes to induce mTORC1 activity
is a characteristic of numerous cancer cells [16, 17]. The
role of GCN2 in tumorigenesis has been studied to a lesser
extent. Several studies have shown that the presence of
GCN2 can contribute to tumor growth in the context of
hypoxia or non-essential amino acid deprivation [18, 19].
Nevertheless, GCN2 has also been shown to contribute to
cell death of cancer cells in a different setting [20]. This
apparent discrepancy reflects the duality of the GCN2/
eIF2α/ATF4 pathway which, as other stress pathways, can
promote either survival or apoptosis [21, 22] depending on
the length and intensity of the stimulus.
In order to understand how a cell can adapt to a long-
term amino acid deprivation, we choose to develop a cell
line able to resist to this condition from mouse embryonic
fibroblasts (MEFs) rather than a tumor cell line. As resistance
to amino acid deprivation mostly emerge from secondary
mutations, as it is the case for resistance to treatment
[23–25], using a specific tumor cell line presenting several
mutations could have biased the development of our model
and rendered the identification of the resistance mechanisms
more difficult. Therefore, MEFs cells were cultured under
selection pressure in a medium containing very low
concentration of amino acids. As in Darwinian selection,
www.impactjournals.com/oncotarget
few cells evolved to develop the capacity to survive in this
environment, each of which grows as a clonal population.
This selection allows us to generate several independent
AADR (Amino Acid Deprivation Resistant) clones that were
able to both survive and proliferate with low concentration
of amino acids.
In these clones, the study of the two main pathways
involved in the maintenance of proteostasis (i.e GCN2/
eIF2α/ATF4 and mTORC1) revealed a down regulation of
ATF4 expression and function. The same characteristic was
also observed in a cancer cell line resistant to amino acid
deprivation. Our study, in the context of a long-term amino
acid deprivation, designed ATF4 not only as pro-apoptotic
factor but especially as a major limiting factor in the process
of acquisition of amino acid deprivation resistance.
RESULTS
Selection and characterization of clones able to
grow in low level of amino acids
In order to generate clones able to survive and
proliferate with very low level of amino acids, we
progressively submitted mouse embryonic fibroblasts
to a decreasing amount of amino acids. At the end of the
selection process, cells were cultured in a so called “2%
medium” in which the concentration of each amino acid
represented 2% of the initial concentration presents in the
control medium (DMEM) (see mat meth for more details).
Three independent clones were selected from different
culture plates for their capacity to proliferate in the 2%
medium. Those clones will be further designated as Amino
Acid Deprivation Resistant clones or AADR clones. When
submitted to the same 2% medium, parental cells exhibit
a weak proliferation during the first 48 hours (Figure
1A) together with a strong proportion of cells with a high
chromatin compaction, an apoptosis marker (Figure 1B).
After 48h of amino acid deficiency, the number of parental
cells decreased. On the contrary, the AADR clones present
a very weak percentage of apoptotic cells. This observation
was corroborated by the study of the cleaved form of caspase
3. Indeed, an increase of this apoptosis marker was observed
after 24 hours of amino acid limitation in the parental cells
whereas no significant signal was observed in the AADR
clones cultured in the same condition (Figure 1C). These
results demonstrate that the AADR clones are not only able
to proliferate in a low concentration of amino acids, but they
have also developed the ability to resist to the apoptotic
process. Strikingly, the AADR clones were also able to
survive to the total absence of amino acids (Figure 1D.),
a drastic situation known to induce cell death [26]. Prior
to further analysis, we have checked whether the clone’s
ability to proliferate and survive in low level of amino acids
was reversible or not. The AADR clones were cultured in
complete medium for two weeks before going back to 2%
medium to test whether their ability to resist to amino acid
deprivation persists. We have shown (Supplementary Figure
27441 Oncotarget
 1) that the AADR clones have kept the capacity to proliferate
(with no latency) in 2% medium thus, confirming that this
feature was acquired in a stable manner.
Activity of the amino acid-regulated pathways in
the AADR Clones
Amino acid availability is known to regulate mainly
two pathways, i.e. the GCN2 and mTORC1 pathways.
These two pathways have been characterized for their
contribution to the control of proteostasis through the
regulation of translation and autophagy during amino
acid deprivation. We thus assessed whether the adaptation
of AADR clones to amino acid deficiency resulted in a
modification of one of these pathways.
In the parental cells, the 2% medium induces the
phosphorylation of GCN2 and of its target eIF2α, [9]
Moreover, an inhibition of S6K1 phosphorylation, a target
of mTORC1 (Figure 2A) was observed in the parental
cells cultured in the 2% medium. These phosphorylation
states of eIF2α and S6K1 are associated with a drastic
inhibition of protein synthesis in 2% medium compared
to control medium as shown in Figure 2B by SUnSet
technique. Likewise, a significant phosphorylation of
GCN2 was also observed in the AADR clones cultured in
the 2% medium (Figure 2A). We also showed that both the
levels of phosphorylation of eIF2α and S6K1 and the level
of protein synthesis were equivalent to those observed in
the parental cells cultured in 2% medium. These results
are consistent with several studies that have highlighted

Figure 1: The AADR clones are resistant to the apoptosis induced by amino acid deprivation. (A) Proliferation assay of
parental cells (Parental) and 3 independent clones (Clone 1, 3, 5) cultured for 24, 48 and 120 hours in 2% medium (Concentration of amino
acid compared to the control medium). Cell numbers were determined by counting. For each cell type, cell number is expressed relative to
the number of cells at 0h. The graph show means ± S.E.M. of 3 independent experiments. Differences were assessed by 1-way ANOVA;
*** indicates a significant difference (p<0,001) between clones and parental cells. (B) Early apoptosis detection using Hoechst 33342
labelling in parental cells and AADR clones cultured for 0, 24 and 48 hours in 2% medium. Results are expressed as the percentage of
cells presenting high Hoechst staining. Graph shows means ± S.E.M. of 3 independent experiments. Differences were assessed by 1-way
ANOVA; *** indicates a significant difference (p<0,001) between clones and parental cells. (C) Parental cells and AADR clones were
cultured in control medium (Ctl) or in 2% medium (2%) for 24h and 48h. Immunoblot analyses of caspase 3 cleavage were performed. (D)
Parental cells (Parental) and AADR clones (Clone 1, 2, 3) were cultured in medium lacking all amino acids for 48 hours and 72 hours. Cell
numbers were determined by counting. For each cell type, cell number is expressed relative to the number of cells at 0h. Graph shows means
± S.E.M. of 3 independent experiments. Differences were assessed by 1-way ANOVA; *** indicates a significant difference (p<0,001)
between parental cells and clones.

www.impactjournals.com/oncotarget 27442 Oncotarget

 Figure 2: The GCN2 and the mTORC1 pathways are still regulated in the AADR clones. Parental cells (Parental) and
AADR clones (Clone 1, 3, 5) were cultured in control medium (Ctl) or in 2% medium (2%) for 24h and 48h (A) Immunoblot analyses
of GCN2, eIF2α, S6K1 and their phosphorylated form were performed. (B) (Upper panel) Protein synthesis was measured by SUnSet
method, immunoblot analyses of puromycin incorporation into proteins were performed. (Lower panel) Puromycin incorporation was
determined by densitometry analysis. Graph shows means ± S.E.M. of 3 independent experiments. Differences were assessed by 1-way
ANOVA; # indicates a significant difference (p<0,001) compared to parental cells in control medium. (C) Parental cells and AADR clones
were cultured for 24h in control medium or in 2% medium. When indicated, 20μM of chloroquine (Cq) was added during the last hour
of treatment. (Upper panel) Immunoblot analyses of eiF2α and LC3B processing (LC3-I and LC3-II) were performed. (Lower panel)
LC3-II level was determined by densitometry analysis and normalized by red ponceau. Graph shows means ± S.E.M. of 3 independent
experiments. Differences were assessed by 1-way ANOVA; *** indicates a significant difference (p<0,001) between clones and parental
cells for a same treatment, # indicates a significant difference (p<0,01) between cells treated or not with chloroquine. (D) Parental cells
and AADR clones were cultured in control medium or in 2% medium for 24h. Map1lc3b mRNA level was determined and normalized by
the level of β-actin mRNA, results are expressed relative to the value observed in parental cells in control medium. Graph shows means
± S.E.M. of 5 independent experiments. Differences were assessed by 1-way ANOVA; # indicates a significant difference (p<0,001)
compared to parental cells in control medium, * indicates a significant difference (p<0,05) between clones and parental cells.

www.impactjournals.com/oncotarget 27443 Oncotarget

 the importance for cells, notably tumor cells, to reduce
protein synthesis level in order to face nutrient restriction
[27]. It also suggests that the decrease of protein synthesis
is not the limiting step concerning the ability to proliferate
upon amino acid scarcity.
These two pathways are also known to regulate
autophagy during amino acid deprivation [28–33]. We
thus checked the level of LC3 lipidation (LC3-II, a marker
of autophagy) in presence or in absence of chloroquine
(an inhibitor of lysosome-autophagosome fusion), to have
an insight into the autophagic flux [34]. In parental cells,
chloroquine addition in 2% medium induced a strong
increase of LC3-II level whereas only a faint increase
was observed in control medium treated with chloroquine
(Figure 2C). This demonstrates that autophagy is induced
in parental cells cultured in 2% medium. As for clones,
strikingly, both LC3-I and II protein levels were weaker
than in the parental cells. The addition of chloroquine
demonstrated that there was a significant autophagy flux
in the AADR clones but apparently reduced compared to
parental cells (Figure 2C).
This experiment does suggest that the difference of
LC3 protein level is not due to a drastic increase in the
autophagic flux but rather to a decrease of its expression.
That was indeed confirmed by measuring the level of the
mRNA encoding LC3 protein. A decrease of around 60%
of the level of the map1Lc3b transcript was observed
in the AADR clones cultured in 2% medium compared
to parental cells (Figure 2D). Interestingly, map1Lc3b
transcription is known to be controlled by ATF4, a
downstream factor of GCN2/eIF2α pathway [35, 36].
Expression of ATF4 and its target genes in the
AADR clones
Considering that ATF4 is involved in the control
of map1Lc3b transcription, we decided to investigate its
protein level in the AADR clones. As expected, in parental
cells, the phosphorylation of eIF2α was associated with
the increase of ATF4 protein level during amino acid
deprivation (Figure 3A). As previously described for
a single essential amino acid deprivation [37], the level

Figure 3: The expression of ATF4 protein and its target gene is diminished in the AADR clones. Parental cells (Parental)
and AADR clones (Clone 1, 2, 3) were cultured in control medium (Ctl) or in 2% medium (2%) for 24h and 48h. (A) (Left panel)
Immunoblot analyses of ATF4 and eIF2α were performed. (Right panel) ATF4 level was determined by densitometry analysis. Graph shows
means ± S.E.M. of 5 independent experiments. Differences were assessed by 1-way ANOVA. # indicates a significant difference (p<0,001)
compared to parental cells in control medium., ** indicates a significant difference (p<0,01) compared to parental cells in 2% medium. (B)
Parental cells and AADR clones were cultured in control medium or in 2% medium for 24h. Asns, Atf3, Atg12 and Trb3 mRNA levels were
determined and normalized by the level of β-actin mRNA, results are expressed relative to the value observed in parental cells in control
medium. Graph show means ± S.E.M. of 5 independent experiments. Differences were assessed by 1-way ANOVA; # indicates a significant
difference (p<0,001) compared to parental cells in control medium, * p<0,05 compared to parental cells in 2% medium.

www.impactjournals.com/oncotarget 27444 Oncotarget

 of Atf4 mRNA was also increased by the 2% medium
(Supplementary Figure 2.). Despite an equivalent level
of eIF2α phosphorylation (cf. Figure 2A), the amount of
ATF4 is significantly lower in the AADR clones compared
to the parental cells cultivated in the 2% medium. In
accordance with this result, we observed that the level
of Atf4 mRNA was also lower in the AADR clones in
comparison with the parental cells (Supplementary Figure
2.). These results illustrate that AADR clones present a
significant lower expression of ATF4 at both protein and
mRNA level.
In order to determine whether this decrease of ATF4
expression is sufficient to impact the ATF4-dependent
gene expression program, we studied, in addition to
map1Lc3b, the mRNA expression of several ATF4-
dependent genes such as Asns, Atf3, Atg12 and Trb3 [32,
38–40]. As expected, the increase of ATF4 expression in
the parental cells cultured in 2% medium was associated
with an important increase in the expression of these
different genes compared to the control condition (Figure
3B). However, for all these genes, the amounts of their
mRNA were significantly lower in the AADR clones
compared to the parental cells cultured in 2% medium.
The alteration of these ATF4-target genes regulation
in AADR clones is coherent with the observation that
these clones present an alteration of ATF4 expression.
To rule out that the decrease in ATF4 expression and its
target genes was a consequence of the long term culture
in 2% medium rather than a modification linked to the
adaption mechanism, we cultured the AADR clones for
two weeks in the control medium before to replace them
in 2% medium for 24 and 48 hours (a protocol similar to
the one applied to the parental cells). This experimental
paradigm shows that the down-regulation of ATF4 and its
target genes is a stable trait acquires during the selection
(Supplementary Figure 3).
Role of the downregulation of ATF4 in
conferring resistance to amino acid deficiency
It was crucial to determine whether this down-
regulation of ATF4, acquired during the selection,
contributes to the adaption to the low amino acid
environment or if it is just a consequence of this adaption.
To assess this question, MEFs cells were infected with a
lentivirus containing a shRNA targeting Atf4 mRNA or a
control shRNA. As expected, the level of Atf4 mRNA and
protein were dramatically decreased in the cells infected
with the shRNA ATF4 cultivated in the 2% medium
compared to the cells infected with the shRNA control
(Figure 4A). This decrease of ATF4 expression resulted
in the decrease of Asns mRNA level, confirming the
functional impact of ATF4 knock-down (Supplementary
Figure 4). Then, the consequence of the down regulation
of ATF4 expression on cell survival in 2% medium was
determined. Knocking-down ATF4 in parental cells
www.impactjournals.com/oncotarget
resulted in an absence of caspase 3 cleavage after 24h and
48h in 2% medium whereas caspase 3 cleavage is observed
in parental cells infected with the control shRNA (Figure
4B). As for proliferation, the absence of ATF4 consecutive
to the infection with Atf4-shRNA enables control cells
to slightly proliferate after 5 and 7 days in 2% medium
whereas the number of cells significantly decreases
after infection with the control shRNA (Figure 4C).
Reciprocally we tested if the overexpression of ATF4 in
the AADR clone would decrease their resistance to amino
acid limitation. The overexpression obtained by infecting
the cells with an adenovirus expressing ATF4 led to a very
variable level of expression of ATF4 between the different
clones (Supplementary Figure 5), this feature represents
a limit for the interpretation of the data. Nevertheless,
for each clones the increase of ATF4 expression was
associated with the cleavage of the caspase 3 and an arrest
of cell growth. Together, these results demonstrate that the
decrease of ATF4 expression increases the ability of cells
to survive during prolonged amino acid deprivation.
ATF4 expression in amino acid deficiency-
resistant cancer cell lines
It is already admitted that ATF4 can contribute to
cell death during prolonged or intense stress [21]. Our
knock-down experiments confirm this concept for MEFs
in the setting of a partial amino acid starvation. We made
the hypothesis that a similar adaptation to amino acid
deprivation involving a low level of expression of ATF4
could be observed in a cancer cell line. We choose to
focus our study on pancreatic cancer cell lines. Indeed,
pancreatic tumors present a high metabolism and a dense
stroma which conduct to an impaired development of the
tumor vasculature and thus, to the development of hypoxic
and nutrient deprived region [41, 42].
We tested two human pancreatic cancer cell lines:
MIA PaCa-2 and BxPC-3. MIA PaCa-2 cells have
been characterized for their poor tolerance to nutrient
deprivation whereas BxPC-3 have been shown to be
more resistant [43]. Based on our previous study, we
also used HeLa cells, a cervical cancer cell line, as a
positive control of ATF4 expression under amino acid
deprivation [37]. Among those cell lines, BxPC-3 cells
exhibited the strongest increase of their cell number after
3 days of culture in 2% medium (Figure 5A). In addition,
relatively to the cell number observed after 3 days in
control medium, the BxPC3 cell line exhibited a decrease
of only 40% in 2% medium (Supplementary Figure 6).
Comparatively, HeLa and MIA PaCa-2 presented a strong
decrease, respectively of 90% and 95% of their cell
numbers. In these two cell lines, the strong inhibition of
cell growth was accompanied by the appearance of the
cleaved form of PARP, a target of caspase-3 (Figure 5C).
In BxPC-3, the cleaved form of PARP was not detected,
suggesting that the 2% medium did not induce apoptosis
27445 Oncotarget
 in this cell line. Remarkably, this capacity of the BxPC-3
to resist to apoptosis was associated with the weaker level
of expression of ATF4 detected among the three cell lines
tested (Figure 5B).
Thus, the capacity of a human tumor cell line to
survive in low amino acid environment could be due to a
decreased expression of ATF4. Therefore, overexpressing
ATF4 in BxPC3 cells may be a good strategy to sensitize
them to the 2% medium. Indeed, overexpression of
ATF4 by adenovirus infection leads to the induction of
apoptosis of BxPC-3 cells in 2% medium as shown by
the detection of the cleaved form of PARP (Figure 6A)
and the significant increase of cells with a high chromatin
compaction (Figure 6B). Moreover, this apoptosis
induction was associated with a reduction of proliferation
compared to BxPC-3 cells infected with Ad-GFP (Figure
6C). All together, these results confirm the hypothesis
that a weak level of expression of ATF4 can contribute
to the capacity of a subset of tumor cells to survive and
proliferate in low concentration of amino acids.
DISCUSSION
The aim of this study was to identify mechanisms of
resistance that can occur in tumor cells due to the pressure
of selection from an amino acid-deprived environment.
For that purpose, we generated clones able to grow in a
medium containing low amount of amino acids (Figure 7.).
It is noticeable that parental cells exposed to a very low
level of amino acids exhibit after 24 hours the initiation of
apoptosis while their growth is not yet totally abolished. In
contrast, the AADR clones were resistant to the apoptosis
induced by a very low concentration of amino acids.
Until now two signaling pathways, mTORC1 and
GCN2/eIF2α/ATF4, are well known to be involved in
the adaption to the variation of amino acid supply. The
mTORC1 pathway did not appear to behave differently
in the AADR clones and the parental cells since in both
cases S6K1 phosphorylation was abolished upon amino
acid limitation. If the inhibition of mTORC1 has been
proposed as a therapeutic strategy to treat cancers, it is

Figure 4: The knock-down of ATF4 expression in MEFs increases cell survival during amino acid deprivation. MEFs
were infected with a lentivirus expressing either a ShRNA control (ShCtl) or a ShRNA targeting ATF4 (ShATF4). (A) ShCtl cells (ShCtl)
and ShATF4 cells (ShATF4) were cultured in control medium (Ctl) or in 2% medium (2%) for 24 hours. (Left panel) Immunoblot analyses
of ATF4 and eIF2α were performed. (Right panel) ATF4 mRNA level was determined and normalized by the level of β-actin mRNA, results
are expressed relative to the value observed in ShCtl cells in control medium. Graph shows means ± S.E.M. of 3 independent experiments.
Differences were assessed by 1-way ANOVA; # indicates a significant difference (p<0,001) compared to control medium *** indicates
a significant difference (p<0,001) compared to ShCtl cells in 2% medium. (B) ShCtl cells and ShATF4 cells were cultured in control
medium or in 2%medium for 24 and 48 hours. Immunoblot analyses of caspase 3 cleavage were performed. (C) ShCtl cells and ShATF4
cells were cultured in 2% medium for 5 days (D5) and 7 days (D7). Cell numbers were determined by counting. For each cell type, cell
number is expressed relative to the number of cells at 0h (represented by the black line). Graph shows means ± S.E.M. of 3 independent
experiments. Differences were assessed by 1-way ANOVA; # indicates a significant difference (p<0,01) compared to ShCtl cells, **
indicates a significant difference (p<0,01) compared to 0h.

www.impactjournals.com/oncotarget 27446 Oncotarget

 now established that its inhibition alone may not be
sufficient to induce apoptosis [44].
Interestingly, we observed an alteration of the eIF2α/
ATF4 pathway in the AADR clones. Indeed, compared to
the parental cells, the level of ATF4 protein was decreased
in the clones despite an equivalent level of eIF2α
phosphorylation. It remains complex to determine the
cause of this low expression level as ATF4 expression is
highly regulated at the level of its transcription, translation
and protein stability [45]. The low amount of ATF4 protein
is associated with a decreased amount of its mRNA. But,
it cannot be concluded that the decrease of Atf4 mRNA
is the initial event contributing to a low ATF4 protein
level as it has been suggested that ATF4 might activate its
own transcription [46]. ATF4 is well known to be highly
regulated at the translational level [47]. Nevertheless,
the levels of both eIF2α phosphorylation and protein
synthesis observed in the AADR clones suggest that the
under-expression of ATF4 is not due to a defect of its
translation. ATF4 expression is also tightly regulated at
the level of protein degradation by the proteasome [45].
This regulation involved the ubiquitin ligase βTrCP, this
enzyme targets notably tumor suppressor proteins for
proteasomal degradation [48]. It will be interesting to
determine whether this process could be involved in the
low level of ATF4 expression in the AADR clones. If
further studies would be necessary to identify the exact
cause of this under expression, it has to be kept in mind
that it may differ for each AADR clone.
The fact that the down regulation of ATF4 and its
target genes are common characteristics of the tested
independent clones suggests that this alteration may
have an important role in the resistance to amino acid
deprivation. This concept has been confirmed by the
knock-down of ATF4 in the parental cells which have
acquired a resistance to the apoptosis induced by amino

Figure 5: The expression of ATF4 is associated to the resistance to amino acid deprivation in the cancer cell line BxPC-
3. (A) HeLa, MIA PaCa-2 and BxPC-3 were cultured in 2% medium (2%) for 72 hours. Cell numbers were determined by counting. For
each cell type, cell number is expressed relative to the number of cells at 0h (represented by the black line). Graph shows means ± S.E.M. of
3 independent experiments. Differences were assessed by 1-way ANOVA; # indicates a significant difference (p<0,001) between cell lines,
** indicates a significant difference (p<0,01) compared to 0h. (B) MIA PaCa-2, BxPC-3 and HeLa cells were cultured control medium or
2% medium for 24 hours. Immunoblot analyses of ATF4, eIF2α and its phosphorylated form and the cleaved form of PARP were performed.

www.impactjournals.com/oncotarget 27447 Oncotarget

 Figure 6: The overexpression of ATF4 in BXPC-3 induces cell death and reduces cell proliferation during amino acid
deprivation. BxPC-3 cells were infected with adenovirus expressing either GFP (Ad-GFP) or mouse form of ATF4 (Ad-ATF4). (A) Cells
infected with either Ad-GFP or Ad-ATF4 were cultured in control medium (Ctl) or in 2% medium (2%) for 24 hours. Immunoblot analyses
of ATF4, eIF2α and the cleaved form of PARP were performed. The arrow indicates the mouse form of ATF4. (B) Early apoptosis detection
using hoechst 33342 labelling in cells infected with either Ad-GFP or Ad-ATF4 cultured in 2% medium for 24 and 48 hours. Results
are expressed as the percentage of cells presenting high hoechst staining. Graph shows means ± S.E.M. of 3 independent experiments.
Differences were assessed by 1-way ANOVA; ** indicates a significant difference (p<0,01) between compared to 0h. (C) Cells infected
with either Ad-GFP or Ad-ATF4 were cultured in 2% medium for 24 and 48 hours. Cell numbers were determined by counting. For each
cell type, cell number is expressed relative to the number of cells at 0h (represented by the black line). Graph shows means ± S.E.M. of
3 independent experiments. Differences were assessed by 1-way ANOVA; # indicates a significant difference (p<0,001) between cells
infected with Ad-GFP and Ad-ATF4, ** indicates a significant difference (p<0,01) compared to 0h.

Figure 7: Summary diagram of the AADR clones. In the short term, the 2% amino acid medium induces rapidly the apoptosis of
mouse embryonic fibroblasts, through notably the induction of ATF4 expression. Nevertheless, this high selective pressure results in the
selection of Amino Acid Deprivation Resistant (AADR) clones. These clones present an alteration of the activation of the GCN2/eIF2α/
ATF4 pathway, despite the phosphorylation of GCN2 and eIF2α, the level of ATF4 remains low, preventing them from apoptosis.

www.impactjournals.com/oncotarget 27448 Oncotarget

 acid deprivation. This pro-apoptotic role of ATF4 is in
accordance with data of the literature showing that the
induction of ATF4 expression, by drugs or glutamine
deprivation [49, 50], has a pivotal role in the initiation of
apoptosis [51].
However, other studies have demonstrated that
ATF4 is necessary for the survival of tumor cells in
deprived environment [18]. ATF4 has been shown to
be necessary for the regulation of VEGF expression in
low vascularized area to promote angiogenesis and so,
the furniture of amino acids [18]. The study of Ye et al.
demonstrates that the presence of ATF4 is necessary to
synthesize non-essential amino acids and more particularly
asparagine [19]. Nevertheless, the authors used for their
in vitro experiments a medium lacking the non-essential
amino acids. Our model differs from that approach as the
scarcity concerns all the amino acids and in particular
the essential ones that cannot be synthetized. These two
examples illustrate the complexity to establish a model
concerning the role of ATF4 in the survival/apoptosis
balance in different contexts of amino acid deprivation.
This complexity resides in the fact that ATF4 increases
the expression of numerous pro-survival and pro-apoptotic
genes. In addition, the capacity to resist to amino acid
scarcity is likely to reside in the modification of expression
of a set of genes rather than in the down regulation of
only one pro-apoptotic gene such as Trb3 or Atf3. Our
data clearly show that ATF4 down regulation protects
from the apoptosis induced by the long-term amino acid
deprivation.
It is reasonable to make the hypothesis that other
mechanisms, independently of ATF4 down regulation,
could contribute to the resistance of the AADR clones.
Another important observation is that the phosphorylation
of eIF2α is conserved in the AADR clones, suggesting
that this event could be necessary to face the amino
acid deprivation. This hypothesis is in accordance with
the study of Han et al. which has demonstrated that the
phosphorylation of eIF2α and the resulting inhibition
of protein synthesis are in favor of survival [52]. The
importance of protein synthesis limitation to resist to long
term amino deprivation is well illustrated by the study
of Leprivier et al [27]. The authors demonstrate that the
decrease of the elongation rate through the activation
of the eukaryotic Elongation Factor 2 Kinase (eEF2k)
promotes cell survival during nutrient deprivation.
Moreover, a high level of eEF2K is associated with
poor diagnostic in different kind of tumors [27]. The
protective effect of eEF2K has been shown to rely on the
inhibition of protein synthesis and not on the induction
of autophagy [53] which has been reported to be induced
by eEF2K [54, 55]. Similarly, in the AADR clones our
results suggest that their capacity to survive in amino acid-
deprived medium do not rely on the increase of autophagy.
Indeed, the levels of autophagic genes were decreased
and the total autophagic flux seems to be decreased.
www.impactjournals.com/oncotarget
As already mentioned, a high level of autophagy can
represent only a transient adaption to a lack of amino
acids. Moreover, it has been demonstrated that a link
exists between autophagy and the induction of apoptosis
[8], consequently, limiting the autophagy may represent
an efficient strategy to resist to a long term amino acid
deprivation.
Finally, we have been able to establish that this
feature of a low level of ATF4 expression leading to amino
acid deprivation resistance occurred also in, at least, one
cancer cell line. Indeed, BxPC-3 pancreatic cancer cell
line exhibited both the capacity to grow in amino acid-
deprived medium and a low level of ATF4 expression.
Moreover, overexpression of ATF4 in BxPC-3 cells re-
sensitizes them to the apoptosis induced by amino acid
deprivation. As cancer cells emerged from a primary
mutation which could be followed by others, it remains
striking to observe the same characteristics of amino
acid deprivation resistance in BxPC-3 and in the AADR
clones. Despite the presence of differential selection
pressures and maybe different mechanisms between
AADR clones and BxPC-3, the decreased expression of
ATF4 seems to be an important characteristic in order for
cell to survive to amino acid deprivation. This observation
establishes that the cellular model that we developed is
pertinent to identify a feature of the resistance to amino
acid deprivation that can be shared by cancer cells.
Importantly, our study does not aim at demonstrating
a strict correlation between ATF4 expression and cell
survival upon amino acid deprivation. As mentioned above
others mechanisms of resistance may arise in tumor cells.
Yet, increasing the level of expression of ATF4 might be
an efficient strategy to target specific tumor cells able to
resist to deprived environment, not only because ATF4 can
be a pro-apoptotic factor but also, because low level of
ATF4 might be the cause of the resistance. For that reason,
the comprehension of the causes of the decrease in ATF4
expression would be of particular importance. It remains
that strategies to target ATF4 expression will have to be
considered in the complexity of tumor heterogeneity and
it is likely that combined or sequential therapies will be
necessary.
MATERIALS AND METHODS
Cell culture and treatment conditions
Mouse embryonic fibroblasts (MEFs) were cultured
at 37°C in humidified 5% CO2 atmosphere in control
medium prepared from Dulbecco’s Modified Eagle’s
Medium (Sigma) containing 10% fetal bovine serum and
supplemented with non-essential amino acid mix (Gibco),
glutamine, penicillin/streptomycin and gentamycin.
HeLa cervical cancer cell line, MIA PaCA-2 and BxPC-
3 pancreatic cancer cell lines were cultured in control
medium. MIA PaCA-2 and BxPC-3 were kindly given
27449 Oncotarget
 by Dr. S. Pyronnet (CRCT, INSERM, Toulouse, France).
Prior to amino acid starvation experiments, cells were
washed twice with phosphate-buffered saline (PBS) and
refed with the appropriate medium.
To obtain cells resistant to amino acid deprivation,
MEFs cells were seeded in several dish plates and
exposed to mediums with decreased concentration of
amino acids. Those mediums were made by diluting the
control medium with DMEM lacking all amino acids and
supplemented with 10% of dialyzed fetal bovine serum.
The concentration of amino acids was gradually decreased
during several weeks causing severe growth arrest and
cell death. The last remaining cells were exposed to the
medium containing only 2% of the amino acids present
in the control medium (medium named 2% medium)
again for several weeks. Few cells survived and formed
individual clones. Three clones were selected form 3
independent plates and were continuously cultured in the
2% medium. The cells from which clones were derived
were thus called parental cells.
Analysis of gene expression using
real time RT-PCR
Total RNA was prepared using RNeasy mini
kit (Qiagen) and treated with DNase I, Amp Grade
(Invitrogen, Carlsbad, CA, USA) prior to cDNA
synthesis. RNA integrity was electrophoretically verified
by ethidium bromide staining. RNA (1 μg) was reverse
transcribed with 100 U of Superscript II plus RNase H-
Reverse Transcriptase (Invitrogen) using 100 μM random
hexamer primers (Amersham Biosciences, Piscataway,
NJ, USA), according to the manufacturer’s instructions.
Real-time quantitative PCR was carried out on a Bio-Rad
CFX-96 detection system with quantitative qPCR SYBR
Green reagents (Bio-Rad, Hercules, CA, USA) and with
a primer concentration of 0.5 μM. For a list of primer
sequences, see Supplementary Table 1. PCR conditions
were standardized to 39 cycles of: 95°C for 08 s, 59°C for
05 s and 72°C for 10 s.
Antibodies
eIF2α, P-eIF2α (Ser51), Total S6K1, P-S6K1
(T389), Cleaved PARP, total and cleaved caspase-3
antibodies were purchased from Cell Signaling
Technologies, ATF4 (SC-200) from Santa Cruz and
MAP1LC3B antibody from Novus. Antibody Puromycin
(clone 12D10) was kindly given by Phillipe Pierre (Centre
d'Immunologie de Marseille-Luminy, France).
Immunoblot analysis
Cells were lysed in a lysis buffer containing (50 mM
Tris, 25 mM β-glycerophosphate, 50 mM NaCl, 1 mM
EDTA, 1 mM EGTA, 0.5% (v/v) Triton X100, 1 mM DTT,
www.impactjournals.com/oncotarget
1 mM benzamidine, protease and phosphatase inhibitor
cocktail (Sigma). Proteins were separated by SDS-
polyacrylamide gel electrophoresis and transferred onto
a Hybond-P PVDF membrane (Amersham Biosciences).
Membranes were blocked for 1 h at room temperature with
a solution of 5% nonfat milk powder in TN (50mM Tris-
HCL, pH 8.0, 150mM NaCl, 0.1% Tween-20). The blots
were then incubated with antibody in blocking solution
overnight at 4°C. Antibodies were diluted according to the
manufacturer’s instructions. The blots were washed three
times in TN and incubated with horseradish peroxidase
conjugated goat anti-IgG (1:5000) (Santa Cruz, CA) in
blocking buffer for 1 h at room temperature. After three
washes, the blots were developed using the LuminataTM
Western HRP substrate (Millipore Billerica, MA).
Establishment of stable cell lines
expressing ShRNA
The lentiviral vectors harboring ATF4 and control
shRNA sequences were obtained from Sigma-Aldrich
(Mission TRC shRNA collections). The lentiviral particles
were produced by EVIR (Enveloppes Virales et Ingénierie
des Rétrovirus, Unité de Virologie Humaine, Inserm
U758, Ecole Normale Supérieure de Lyon, France). MEFs
cells were seeded in 6 wells plates and infected the next
day with lentivirus expressing either a control ShRNA or
the ShRNA targeting ATF4 at a Multiplicity of Infection
(MOI) of 25. Twenty four hours after the infection, cells
were washed twice and refed with DMEM medium
containing the selective agent puromycin.
Hoechst 33342 staining assay/Fluorescent
microscopy evaluation of cell apoptosis and
morphology
Cells were stained with 2 μg/mL of Hoechst 33342
for 5 min at 37°C. Then, cells were washed twice with
PBS and refed with amino acid-deprived medium. Hoechst
33342 fluorescence was assessed using Zen microscope
(Zeiss) with DAPI filter (Excitation 353 nm/ Emission
465 nm). Early apoptotic cells were detected by strong
Hoechst staining which represented the condensation of
nuclear chromatin and its fragmentation. For each cell
types and conditions, several random areas were selected
for a total count of approximatively 300 cells. The number
of apoptotic cells was normalized to the total number of
cells in each area.
Cell proliferation assays
Cells were plated in 12-wells plates and, the next
day, cultured in either control medium or in amino acid-
deprived medium. The proliferation rate of each cell types
was assessed by counting 3 wells for each time points
using Beckman coulter counter Z1. The number of cells
27450 Oncotarget
 was normalized according to the number of cells counted
at the beginning of the experiment.
Adenoviral infection
Cells were plated the day prior infection. The
next day, cells were washed twice and refed with
control medium without FBS. Infection with adenovirus
expressing either GFP or mouse ATF4 was carried at a
multiplicity of infection of 100. After 8h of infection,
FBS was added to the medium for a final concentration of
10%. After 24h of infection, the cells were washed twice
and refed with either control medium or 2% medium.
Adenovirus were kindly given by Guozhi Xiao (Rush
University Medical Center, Chicago, USA).
Measure of protein synthesis level by
SUnSet method
The level of protein synthesis has been determined
by the SUnSet method [56]. Briefly, during the last 30
min of treatment puromycin (Sigma-Aldrich) at 5μg/ml
was added to the culture medium. Cells were harvested
and lysed for immunoblot analysis with anti puromycin
antibody.
Statistical analysis
All data are expressed as means ± SEM. Differences
between samples were assessed by one-way ANOVA
(fisher test).
ACKNOWLEDGMENTS
This work was supported by grants from the Institut
National de la Recherche Agronomique (INRA) and from
Fondation ARC pour la Recherche sur le Cancer. F. Mesclon
was supported by INRA and Auvergne-Rhône-Alpes grant.
CONFLICTS OF INTEREST
The authors declare that they have no conflicts of
interest.
REFERENCES
1. Vaupel P, Kallinowski F, Okunieff P. Blood Flow, Oxygen
and Nutrient Supply, and Metabolic Microenvironment of
Human Tumors: A Review. Cancer Res. 1989; 49: 6449–65.
2. Graeber TG, Osmanian C, Jacks T, Housman DE, Koch
CJ, Lowe SW, Giaccia AJ. Hypoxia-mediated selection of
cells with diminished apoptotic potential in solid tumours.
Nature. 1996; 379: 88–91.
3. Bristow RG, Hill RP. Hypoxia and metabolism: Hypoxia,
DNA repair and genetic instability. Nat Rev Cancer. 2008;
8: 180–92.
www.impactjournals.com/oncotarget
4. Pouysségur J, Dayan F, Mazure NM. Hypoxia signalling
in cancer and approaches to enforce tumour regression.
Nature. 2006; 441: 437–43.
5. Harris AL. Hypoxia — a key regulatory factor in tumour
growth. Nat Rev Cancer. 2002; 2: 38–47.
6. Muñoz-Pinedo C, El Mjiyad N, Ricci J-E. Cancer
metabolism: current perspectives and future directions. Cell
Death Dis. 2012; 3: e248.
7. Geck RC, Toker A. Nonessential amino acid metabolism in
breast cancer. Adv Biol Regul. 2016; 62:11-17.
8. B’Chir W, Chaveroux C, Carraro V, Averous J, Maurin AC,
Jousse C, Muranishi Y, Parry L, Fafournoux P, Bruhat A.
Dual role for CHOP in the crosstalk between autophagy and
apoptosis to determine cell fate in response to amino acid
deprivation. Cell Signal. 2014; 26: 1385–91.
9. Berlanga JJ, Santoyo J, De Haro C. Characterization of a
mammalian homolog of the GCN2 eukaryotic initiation factor
2alpha kinase. Eur J Biochem FEBS. 1999; 265: 754–62.
10. Romano PR, Garcia-Barrio MT, Zhang X, Wang Q, Taylor
DR, Zhang F, Herring C, Mathews MB, Qin J, Hinnebusch
AG. Autophosphorylation in the activation loop is required
for full kinase activity in vivo of human and yeast eukaryotic
initiation factor 2alpha kinases PKR and GCN2. Mol Cell
Biol. 1998; 18: 2282–97.
11. Vattem KM, Wek RC. Reinitiation involving upstream
ORFs regulates ATF4 mRNA translation in mammalian
cells. Proc Natl Acad Sci USA. 2004; 101: 11269–74.
12. Deval C, Chaveroux C, Maurin AC, Cherasse Y, Parry
L, Carraro V, Milenkovic D, Ferrara M, Bruhat A,
Jousse C, Fafournoux P. Amino acid limitation regulates
the expression of genes involved in several specific
biological processes through GCN2-dependent and GCN2-
independent pathways. Febs J. 2009; 276: 707–18.
13. Burnett PE, Barrow RK, Cohen NA, Snyder SH, Sabatini
DM. RAFT1 phosphorylation of the translational regulators
p70 S6 kinase and 4E-BP1. Proc Natl Acad Sci U S A.
1998; 95: 1432–7.
14. Guertin DA, Sabatini DM. An expanding role for mTOR in
cancer. Trends Mol Med. 2005; 11: 353–61.
15. Tsun Z-Y, Possemato R. Amino acid management in cancer.
Semin Cell Dev Biol. 2015; 43: 22–32.
16. Cormerais Y, Giuliano S, LeFloch R, Front B, Durivault
J, Tambutté E, Massard P-A, Ballina LR de la, Endou H,
Wempe MF, Palacin M, Parks SK, Pouyssegur J. Genetic
disruption of the multifunctional CD98/LAT1 complex
demonstrates the key role of essential amino acid transport
in the control of mTORC1 and tumor growth. Cancer Res.
2016; 1;76:4481-92.
17. van Geldermalsen M, Wang Q, Nagarajah R, Marshall AD,
Thoeng A, Gao D, Ritchie W, Feng Y, Bailey CG, Deng N,
Harvey K, Beith JM, Selinger CI, et al. ASCT2/SLC1A5
controls glutamine uptake and tumour growth in triple-
negative basal-like breast cancer. Oncogene. 2016; 35:
3201–8.
27451 Oncotarget
 18. Wang Y, Ning Y, Alam GN, Jankowski BM, Dong Z, Nör
JE, Polverini PJ. Amino Acid Deprivation Promotes Tumor
Angiogenesis through the GCN2/ATF4 Pathway. Neoplasia.
2013; 15: 989.
19. Ye J, Kumanova M, Hart LS, Sloane K, Zhang H, De
Panis DN, Bobrovnikova-Marjon E, Diehl JA, Ron D,
Koumenis C. The GCN2-ATF4 pathway is critical for
tumour cell survival and proliferation in response to nutrient
deprivation. EMBO J. 2010; 29: 2082–96.
20. Wei C, Lin M, Jinjun B, Su F, Dan C, Yan C, Jie Y, Jin
Z, Zi-Chun H, Wu Y. Involvement of general control
nonderepressible kinase 2 in cancer cell apoptosis by
posttranslational mechanisms. Mol Biol Cell. 2015; 26:
1044–57.
21. Hetz C. The unfolded protein response: controlling cell fate
decisions under ER stress and beyond. Nat Rev Mol Cell
Biol. 2012; 13: 89–102.
22. Mazure NM, Pouysségur J. Hypoxia-induced autophagy:
cell death or cell survival? Curr Opin Cell Biol. 2010; 22:
177–80.
23. Sakai W, Swisher EM, Karlan BY, Agarwal MK, Higgins
J, Friedman C, Villegas E, Jacquemont C, Farrugia DJ,
Couch FJ, Urban N, Taniguchi T. Secondary mutations as
a mechanism of cisplatin resistance in BRCA2-mutated
cancers. Nature. 2008; 451: 1116–20.
24. Daver N, Cortes J, Ravandi F, Patel KP, Burger JA,
Konopleva M, Kantarjian H. Secondary mutations as
mediators of resistance to targeted therapy in leukemia.
Blood. 2015; 125: 3236–45.
25. Gibaja V, Shen F, Harari J, Korn J, Ruddy D, Saenz-Vash
V, Zhai H, Rejtar T, Paris CG, Yu Z, Lira M, King D, Qi
W, et al. Development of secondary mutations in wild-type
and mutant EZH2 alleles cooperates to confer resistance to
EZH2 inhibitors. Oncogene. 2016; 35: 558–66.
26. Sadasivan S, Waghray A, Larner SF, Jr WAD, Hayes RL,
Wang KKW. Amino acid starvation induced autophagic cell
death in PC-12 cells: Evidence for activation of caspase-3
but not calpain-1. Apoptosis. 2006; 11: 1573–82.
27. Leprivier G, Remke M, Rotblat B, Dubuc A, Mateo
A-RF, Kool M, Agnihotri S, El-Naggar A, Yu B, Prakash
Somasekharan S, Faubert B, Bridon G, Tognon CE, et
al. The eEF2 Kinase Confers Resistance to Nutrient
Deprivation by Blocking Translation Elongation. Cell.
2013; 153: 1064–79.
28. Blommaart EFC, Luiken JJFP, Blommaart PJE, Woerkom
GM van, Meijer AJ. Phosphorylation of Ribosomal Protein
S6 Is Inhibitory for Autophagy in Isolated Rat Hepatocytes.
J Biol Chem. 1995; 270: 2320–6.
29. Talloczy Z, Jiang W, Virgin HW th, Leib DA, Scheuner
D, Kaufman RJ, Eskelinen EL, Levine B. Regulation of
starvation- and virus-induced autophagy by the eIF2alpha
kinase signaling pathway. Proc Natl Acad Sci USA. 2002;
99: 190–5.
www.impactjournals.com/oncotarget
30. Hosokawa N, Hara T, Kaizuka T, Kishi C, Takamura A,
Miura Y, Iemura S, Natsume T, Takehana K, Yamada N,
Guan JL, Oshiro N, Mizushima N. Nutrient-dependent
mTORC1 association with the ULK1-Atg13-FIP200
complex required for autophagy. Mol Biol Cell. 2009; 20:
1981–91.
31. Jung CH, Jun CB, Ro SH, Kim YM, Otto NM, Cao J,
Kundu M, Kim DH. ULK-Atg13-FIP200 complexes
mediate mTOR signaling to the autophagy machinery. Mol
Biol Cell. 2009; 20: 1992–2003.
32. B’chir W, Maurin A-C, Carraro V, Averous J, Jousse C,
Muranishi Y, Parry L, Stepien G, Fafournoux P, Bruhat A.
The eIF2α/ATF4 pathway is essential for stress-induced
autophagy gene expression. Nucleic Acids Res. 2013; 41:
7683–99.
33. Meijer AJ, Lorin S, Blommaart EF, Codogno P. Regulation
of autophagy by amino acids and MTOR-dependent signal
transduction. Amino Acids. 2015; 47:2037-63.
34. Klionsky DJ, Abdelmohsen K, Abe A, Abedin MJ,
Abeliovich H, Acevedo Arozena A, Adachi H, Adams CM,
Adams PD, Adeli K, Adhihetty PJ, Adler SG, Agam G, et
al. Guidelines for the use and interpretation of assays for
monitoring autophagy (3rd edition). Autophagy. 2016; 12:
1–222.
35. Rzymski T, Milani M, Pike L, Buffa F, Mellor HR,
Winchester L, Pires I, Hammond E, Ragoussis I, Harris
AL. Regulation of autophagy by ATF4 in response to severe
hypoxia. Oncogene. 2010; 29: 4424–35.
36. Rouschop KM, van den Beucken T, Dubois L, Niessen H,
Bussink J, Savelkouls K, Keulers T, Mujcic H, Landuyt W,
Voncken JW, Lambin P, van der Kogel AJ, Koritzinsky M,
et al. The unfolded protein response protects human tumor
cells during hypoxia through regulation of the autophagy
genes MAP1LC3B and ATG5. J Clin Invest. 2010; 120:
127–41.
37. Averous J, Bruhat A, Jousse C, Carraro V, Thiel G,
Fafournoux P. Induction of CHOP expression by amino
acid limitation requires both ATF4 expression and ATF2
phosphorylation. J Biol Chem. 2004; 279: 5288–97.
38. Ohoka N, Yoshii S, Hattori T, Onozaki K, Hayashi H.
TRB3, a novel ER stress-inducible gene, is induced via
ATF4-CHOP pathway and is involved in cell death. EMBO
J. 2005; 24: 1243–55.
39. Jiang H-Y, Wek SA, McGrath BC, Lu D, Hai T, Harding
HP, Wang X, Ron D, Cavener DR, Wek RC. Activating
transcription factor 3 is integral to the eukaryotic initiation
factor 2 kinase stress response. Mol Cell Biol. 2004; 24:
1365–77.
40. Siu F, Bain PJ, LeBlanc-Chaffin R, Chen H, Kilberg MS.
ATF4 Is a Mediator of the Nutrient-sensing Response
Pathway That Activates the Human Asparagine Synthetase
Gene. J Biol Chem. 2002; 277: 24120–7.
41. Guillaumond F, Leca J, Olivares O, Lavaut M-N, Vidal N,
Berthezène P, Dusetti NJ, Loncle C, Calvo E, Turrini O,
27452 Oncotarget
 Iovanna JL, Tomasini R, Vasseur S. Strengthened glycolysis
under hypoxia supports tumor symbiosis and hexosamine
biosynthesis in pancreatic adenocarcinoma. Proc Natl Acad
Sci. 2013; 110: 3919–24.
42. Koong AC, Mehta VK, Le QT, Fisher GA, Terris DJ, Brown
JM, Bastidas AJ, Vierra M. Pancreatic tumors show high
levels of hypoxia. Int J Radiat Oncol. 2000; 48: 919–22.
43. Izuishi K, Kato K, Ogura T, Kinoshita T, Esumi H.
Remarkable Tolerance of Tumor Cells to Nutrient
Deprivation: Possible New Biochemical Target for Cancer
Therapy. Cancer Res. 2000; 60: 6201–7.
44. Benjamin D, Colombi M, Moroni C, Hall MN. Rapamycin
passes the torch: a new generation of mTOR inhibitors. Nat
Rev Drug Discov. 2011; 10: 868–80.
45. Lassot I, Ségéral E, Berlioz-Torrent C, Durand H,
Groussin L, Hai T, Benarous R, Margottin-Goguet F.
ATF4 Degradation Relies on a Phosphorylation-Dependent
Interaction with the SCFβTrCPUbiquitin Ligase. Mol Cell
Biol. 2001; 21: 2192–202.
46. Cho H, Wu M, Zhang L, Thompson R, Nath A, Chan
C. Signaling dynamics of palmitate-induced ER stress
responses mediated by ATF4 in HepG2 cells. BMC Syst
Biol. 2013; 7: 9.
47. Harding HP, Novoa I, Zhang Y, Zeng H, Wek R, Schapira
M, Ron D. Regulated Translation Initiation Controls Stress-
Induced Gene Expression in Mammalian Cells. Mol Cell.
2000; 6: 1099–108.
48. Frescas D, Pagano M. Deregulated proteolysis by the F-box
proteins SKP2 and β-TrCP: tipping the scales of cancer. Nat
Rev Cancer. 2008; 8: 438–49.
www.impactjournals.com/oncotarget
49. Iurlaro R, Muñoz-Pinedo C. Cell death induced by
endoplasmic reticulum stress. FEBS J. 2016; 283:2640-52.
50. Qing G, Li B, Vu A, Skuli N, Walton ZE, Liu X, Mayes PA,
Wise DR, Thompson CB, Maris JM, Hogarty MD, Simon
MC. ATF4 Regulates MYC-Mediated Neuroblastoma Cell
Death upon Glutamine Deprivation. Cancer Cell. 2012; 22:
631–44.
51. Wek RC, Jiang H-Y, Anthony TG. Coping with stress: eIF2
kinases and translational control. Biochem Soc Trans. 2006;
34: 7–11.
52. Han J, Back SH, Hur J, Lin Y-H, Gildersleeve R, Shan
J, Yuan CL, Krokowski D, Wang S, Hatzoglou M,
Kilberg MS, Sartor MA, Kaufman RJ. ER-stress-induced
transcriptional regulation increases protein synthesis
leading to cell death. Nat Cell Biol. 2013; 15: 481–90.
53. Moore CEJ, Wang X, Xie J, Pickford J, Barron J, Regufe da
Mota S, Versele M, Proud CG. Elongation factor 2 kinase
promotes cell survival by inhibiting protein synthesis
without inducing autophagy. Cell Signal. 2016; 28: 284–93.
54. Wu H, Yang J-M, Jin S, Zhang H, Hait WN. Elongation
Factor-2 Kinase Regulates Autophagy in Human
Glioblastoma Cells. Cancer Res. 2006; 66: 3015–23.
55. Py BF, Boyce M, Yuan J. A critical role of eEF-2K in
mediating autophagy in response to multiple cellular
stresses. Autophagy. 2009; 5: 393–6.
56. Goodman CA, Mabrey DM, Frey JW, Miu MH, Schmidt
EK, Pierre P, Hornberger TA. Novel insights into the
regulation of skeletal muscle protein synthesis as revealed
by a new nonradioactive in vivo technique. FASEB J. 2011;
25: 1028–39.
27453 Oncotarget